摘要:
A node of an LSP may inform the ingress node of the LSP, for example via RSVP signaling, about its temporary unavailability for a certain time. In response, the ingress node can stop using any affected LSP(s) and divert the traffic to other LSPs. This provides a faster mechanism to signal traffic shift then traditional IGP overload which causes considerable churn into the network as all the nodes need to compute the SPF. It is sufficient for ingress node to be aware of this node maintenance and it can use information to divert the traffic to other equal cost multipath (ECMP) LSP(s), or other available LSP(s). If no alternative LSP path exists when the ingress node receives such a message, a new LSP can be built during this time and traffic diverted smoothly (e.g., in a make-before-break manner) before the node goes offline for maintenance. Since only the ingress node is responsible to push the traffic to the LSP, there is no need to tear down the LSP for such node maintenance (especially when they are for a short duration). This can be used with a controller responsible for the LSP as well.
摘要:
A network device may receive RSVP path request messages from an upstream ingress network device, and may generate an RSVP path error message, with an overload error code and a timeout period, after the network device is online within the configured timeout period. The network device may provide the RSVP path error message to the ingress network devices to cause the ingress network devices to wait for expiration of the timeout period, after the network device is fully online, until resending the RSVP path request messages. The network device may receive new RSVP path request messages from the ingress network devices after expiration of the timeout period after the network device is fully online, and may establish, based on the new RSVP path request messages, label-switched paths from the ingress network devices.
摘要:
In one example, a method includes receiving, by a network device, first data defining a group of LSPs, receiving second data defining one or more constraints for one or more bypass LSPs, and receiving third data defining a mapping between the group of LSPs and the one or more bypass LSPs. The method also includes, in response to receiving the third data, automatically signaling, by the network device, a bypass LSP in accordance with the one or more constraints, selecting, by the network device and based on the mapping, a respective alternate next hop for rerouting network traffic received on each LSP of the group of LSPs to the signaled bypass LSP, and programming a forwarding component of the network device to install each of the respective alternate next hops as alternate next hops to primary next hops for the LSPs of the group of LSPs.
摘要:
Techniques are described for enhancements to Protocol Independent Multicast (PIM) to support multicast only fast re-route (MoFRR) over a remote loop free alternate (RLFA) backup path in a network. This disclosure describes a modified PIM control message having a new PIM message type and an additional field to indicate an address of a RLFA network device in the RLFA backup path. According to techniques of this disclosure, network devices along the RLFA backup path are configured to forward the modified PIM control message toward the RLFA network device instead of toward a source of a requested multicast group. When the RLFA network device receives the modified PIM control message, the RLFA network device is configured to forward a conventional PIM control message towards the source of the requested multicast group. In this way, PIM can be used to provide MoFRR over a RLFA backup path.